27 replies to this topic

[Andrew Davie]

This project is dead.
I don't have the time or inclination to push it over the line, so rather than leaving people hoping I'm calling an official end to it.
I may release the source code, sans any Boulder Dash content, at some future date.
Cheers
A

[remowilliams]

Aw crap...

[EarthQuake]

Gah. It looked so promising...

[Nathan Strum]

Can't say I'm surprised. Disappointed, but not surprised.

[cd-w]

That is a shame - it seemed to be very close to completion and the copyright issues were resolved? Hopefully someone else can step in and complete the game, or finish one of the variants (e.g. Ruby Runner).

Chris

[iwan-iwanowitsch-goratschin]

That´s tragic!

[Emehr]

Disappointed

I think I just got a soft-on.

[birdie3]

That sucks, man. You definitely do some amazing work though... Could you possibly re-work the engine into a similar yet more original game?

[potatohead]

Bummer. Thanks though!

I enjoyed seeing the tech realized. The VCS can do this game, and that's sweet.

No worries here!

[Dino]

NNNNNNNNNNOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO!!!!!!!!!!!!!!!

[iwan-iwanowitsch-goratschin]

I have played the game in 2008, its development was advanced and it was just awesome to see a game like this
working on the 2600!

Isn´t there any chance that Thomas J. can overtake the project????

[tremoloman2006]

Man this really sux! I was so psyched for this title but I understand Andrew has higher priorities.

I hope someone can take this project over and complete it. God I wish I knew how to program for the 2600!

[STGuy1040]

Damn! I just heard about this project too.

[Thomas Jentzsch]

iwan-iwanowitsch-goratschin, on Tue Dec 16, 2008 8:54 AM, said:

Isn´t there any chance that Thomas J. can overtake the project????

I seriously doubt it. While I got the Ok from Andrew, there is a lot of code which is not completely finished and which I do not know well enough. So that would require a lot of effort to even get into it. Finishing the game alone requires too much of my limited time. And a major part of the fun was cooperating with Andrew, competing for the best solution.

Also I am not interested at all in doing again all that legal stuff Andrew had solved already. 

I am really sad this game didn't come through. But I have a very, very vague hope Andrew may eventually (in a few months or years) decide differently.

[Crazy Climber]

Oh well, I'm sure I'll find a home for the $50 I was going to spend on it. When is that damn AA store sale anyways

[batari]

Thomas Jentzsch, on Tue Dec 16, 2008 1:06 PM, said:

Also I am not interested at all in doing again all that legal stuff Andrew had solved already. 

Why would you need to redo the legal stuff?

[Thomas Jentzsch]

batari, on Wed Dec 17, 2008 11:19 PM, said:

Thomas Jentzsch, on Tue Dec 16, 2008 1:06 PM, said:

Also I am not interested at all in doing again all that legal stuff Andrew had solved already. 

Why would you need to redo the legal stuff?

Because only Andrew got permission, not me. Strictly legally he wasn't even allowed to cooperate with me.

[Propane13]

Would it be considered inappropriate to ask how far along it got (i.e. percentage of completed levels or features)?

I'm also wondering if there was a major technical hangup somewhere that made things stop, or if life just got busy.
Technically speaking, I thought the amoeba would have been the hardest part, and since it got finished, that was the most astonishing tech accomplishment I had seen on the 2600. The recursion and math involved for that would have been astounding.

For the others on this thread, I do have hope as well; I first started "Steam Tunnel Bob" in 2000, and now it's 2009. Maybe it'll get finished by 2010.
But, people need to realize that sometimes, projects have to get buried, and then come back again. Programmers have lives too. Heck, I took some time off to work on my dating life a few years ago, and it was totally worth it.

Anyway, thanks to Andrew and Thomas for the demos that have been posted thus far-- it was a fun realization to see happening, and I am very glad that someone finally went through the legal trouble to make a game "official". As far as I know, this was the first time it had happened, and it was a good inspiration to the community, that such things can be done.

Cheers, Mister Davie. Cheers.
-John

[Thomas Jentzsch]

Well, Andrew asked me about the latest source code a few days ago. Whatever that means...

IIRC the levels where all defined completely, some objects needed a bit of work. What was not completed was the overall framework (scoring, timers, level progress, options etc.). Also we were still tuning the game engine to avoid slowdowns. Then there were some bugs and last not least we were wondering about a different, less restricted bankswitching/RAM board.

So maybe 2/3rd were done.

[Andrew Davie]

Thomas Jentzsch, on Sat Jan 3, 2009 9:22 AM, said:

Well, Andrew asked me about the latest source code a few days ago. Whatever that means...

IIRC the levels where all defined completely, some objects needed a bit of work. What was not completed was the overall framework (scoring, timers, level progress, options etc.). Also we were still tuning the game engine to avoid slowdowns. Then there were some bugs and last not least we were wondering about a different, less restricted bankswitching/RAM board.

So maybe 2/3rd were done.

Yes, the levels are all defined. The major 'bug' is that the creatures (butterflies, boulders, etc) are processed in a different order than the original, and this affects the physics -- sometimes things behave slightly differently. It's not particularly noticeable, but it's not exactly the same as the original - more like 99.5% the same. That's because the original scans a board left to right top to bottom and does things when it finds particular objects at each board location. The 2600 version doesn't have time to do that scan, so all objects are managed in a stack. The stack isn't sorted, so objects tend to get processed in a different order. It shouldn't be impossible to fix -- just sort the stack before processing.

Scoring is done, timers are done, level progress is done.

A title/selection screen needs doing. Also intermission screens.

There are slowdowns, probably unavoidable given the current bankswitch scheme (3E). Switching to a better scheme (4A50) would probably (at a guess) speed up the game 20% but unfortunately that's not going to happen. The engine is rather complex at this stage and much of it I've forgotten (although it's reasonably well commented). I don't have any plans to do a conversion.

This effort was more a proof of concept -- it was for me a kind of 'holy grail' for Atari 2600 programming. I've proved, I think (along with Thomas' help) that it CAN be done, and it can be done well. The major achievements were the 'character' based display system, the 'colour' display consisting of alternating colour bands, and the implementation of a system that actually allowed the game to be played at a decent speed. Also impressive (to me) were the amoeba functionality, which I thought about for a long time, and the magic wall which ended up costing almost no processing time at all.

It's been an interesting, sometimes enjoyable, sometimes very stressful, project. I pick it up now and then and have a play.

Cheers
A

[deathtrappomegranate]

Andrew Davie, on Fri Jan 2, 2009 10:52 PM, said:

This effort was more a proof of concept -- it was for me a kind of 'holy grail' for Atari 2600 programming.

This game is iconic, and the idea of having a 2600 version was a "holy grail" to many people here, I think.

I can't have any sensible discussion about the intricacies of 2600 coding, but I am amazed at what can be squeezed out of that little box by those who really know what they're doing. I am very sorry to hear that this project seems to have reached a dead end.

I guess that a few days after Christmas is the wrong time to be hoping for a resurrection...

[batari]

IIRC, the 3E scheme supports up to 512k ROM/32k RAM, or something like that. How much of each are you actually using?

[Thomas Jentzsch]

batari, on Sat Jan 3, 2009 9:57 AM, said:

IIRC, the 3E scheme supports up to 512k ROM/32k RAM, or something like that. How much of each are you actually using?

I am not sure, 32k ROM + 16k RAM IIRC.

But the biggest problem is the very limited size of the banks. So you soon run out of space and start shifting around stuff which is pretty annoying. Supercat's bankswitching looked much more flexible. But Andrew was the one, who invested time into that, so he knows much better.

[batari]

Thomas Jentzsch, on Sat Jan 3, 2009 4:46 AM, said:

batari, on Sat Jan 3, 2009 9:57 AM, said:

IIRC, the 3E scheme supports up to 512k ROM/32k RAM, or something like that. How much of each are you actually using?

I am not sure, 32k ROM + 16k RAM IIRC.

But the biggest problem is the very limited size of the banks. So you soon run out of space and start shifting around stuff which is pretty annoying. Supercat's bankswitching looked much more flexible. But Andrew was the one, who invested time into that, so he knows much better.

The "magic writes" in 4A50 are what makes it shine, and would certainly make things better for you. However, if Andrew passes the torch to another programmer, and this programmer wants to use 3E, and can live with about 31k ROM/7k RAM, one could target the Harmony cart as a release platform. Also, since the Harmony cart uses a microcontroller, it may also be able to take over some of the processing for the game, like the stack sorting.

[Thomas Jentzsch]

batari, on Sat Jan 3, 2009 9:02 PM, said:

The "magic writes" in 4A50 are what makes it shine, and would certainly make things better for you.

Can you explain those magic writes for me, please?

Quote

However, if Andrew passes the torch to another programmer, and this programmer wants to use 3E, and can live with about 31k ROM/7k RAM, one could target the Harmony cart as a release platform.

7k RAM won't be enough, since the game heavily relies on self-modfying code.